A Methodology of Determining Weight Ratios of Different Question Types Based on Back Propagation Neural Network

This paper introduces an efficient method of determining weight ratios of different question types in the processing of test paper generation based on Back Propagation Neural Network, which will be used in the theory part of vocational skills online tests, which are prepared for staff in power substations. There are 3 types of test questions in the test question bank. The processing of choosing proper ratio of 3 types of test questions for different staff of different skill levels is presented in this paper, which is the key point of generating test paper automatically and reasonably. The ratios gained from this methodology are based on the test results of previous candidates.